SECTION II THE TOOLS

Chapter 3 Biodiversity assessment tools David Allen, Anna McIvor, Channa Bambaradeniya and William Darwall

Species are the components of ecosystems, and their use presence and abundance within freshwater wetlands. It underpins many rural livelihoods. This chapter gives practical presents survey methods for some key freshwater taxa guidelines and approaches for sampling biodiversity (including fi sh, plants, molluscs, and dragonfl ies).

B1 Overview 32 B2 Planning a fi eld survey 34 B3 Conducting species surveys 35 B4 Fish survey sampling methods 36 B5 Mollusc sampling methods 38 B6 Dragonfl y and damselfl y sampling methods 41 B7 Sampling methods for non-fi sh vertebrates associated with wetlands: amphibians, birds and mammals 43 B8 Plant survey methods 45 B9 Market surveys 47 B10 Assessing threats to freshwater species and ecosystems 48 B11 Assessing the conservation status of species 49 B12 Alternative methods for biodiversity assessment 55 Further reading 56

David Allen/Darwin Integrated Wetland Assessment Project

CHAPTER 3 31

IUCN Darwin pp31-56 Section II - Chapter 3.indd 31 16/7/09 15:36:12 SECTION II THE TOOLS

3 Biodiversity assessment tools are made. They have come to be used as the common currency to express biodiversity. Data for species status (such as the IUCN This section presents the biodiversity assessment tools. It Red List of Threatened Species) tend to be more readily available includes: on the global scale, especially for those more charismatic taxa. ✔ An overview of the need for biodiversity assessment, and the Ecosystems may also be used as a measure of biodiversity but, in methods used particular for wetland systems, they remain poorly classifi ed and ✔ Guidance on planning and conducting biodiversity surveys mapped. ✔ Species-specifi c biodiversity sampling methods ✔ Assessment of threats and conservation status of freshwater Wetland biodiversity provides enormous direct and indirect species and ecosystems benefi ts to people. Provisioning services from wetlands, such ✔ Alternative methods for biodiversity assessment as nutrition (notably fi sh) and fi bre are essential for human well- being. Inland fi sheries in developing countries often provide the For collecting information on physical wetland characteristics primary source of animal protein for rural communities as well we recommend an initial literature review is used to swiftly as a vital source of income in many cases, and fl ood plains gather contextual information. provide important grazing for many pastoralists. Supporting and regulating services (such as nutrient cycling) are critical to The physical wetland characteristics set the context for the sustaining ecosystem functions that deliver many benefi ts to habitats, ecosystem services, species and livelihoods that we people (MEA 2005). Wetland ecosystems also play an important will assess through an integrated wetland assessment. role in the regulation of global climate change by sequestering and releasing signifi cant amounts of carbon, as well as providing ✔ What are the conditions that sustain the wetland ecosystem? many other functions locally, regionally and internationally. ✔ How do species, ecosystems and livelihood strategies depend on physical characteristics? Despite the clearly recognised benefi ts provided by wetlands they ✔ What threats, such as global climate change or hydrological continue to be lost at an unprecedented rate and their constituent disturbance, impact upon the sustainability of the system? species are thought more threatened than any other species grouping (see, for example, Ricciardi and Rasmussen 1999, In order to understand the habitat for wetland ecosystems, Revenga et al. 2005). The main threats to global freshwater species general contextual data are needed on the following issues include: overexploitation (of species and water resources); water relating to the study site: pollution; fl ow modifi cation; destruction or degradation of habitat; ● Topography (river basin) and geology (e.g. geological strata and invasion by exotic species (Dudgeon et al. 2006). Pollution and soil types) problems are pandemic. Flow modifi cations are ubiquitous ● Climate and climate cycles in running waters, most often in regions with highly variable ● Hydrological regime – seasonal fl ows and hydrology fl ow regimes where people have the greatest need for fl ood ● Sedimentation protection and water storage (e.g. water storage dams), but also ● Vegetation types as a result of hydroelectric generation through the construction of dams. Habitat degradation is brought about by an array of Data can be collated during the preliminary literature review interacting factors such as conversion for agriculture, pollution, stage of the assessment. Although detailed data may not be forest clearance and resultant changes in surface run-off and available for a particular assessment area, it will usually be general wetland drainage and water abstraction. Invasions by available for river basins or at other sub-national scales from exotic species change the ecological balance through predation, government departments or universities. disease, competition and, in some cases, habitat destruction. The high degree of connectivity within aquatic systems often means B1 Overview that impacts such as pollution or invasive species spread far more rapidly than is usual in terrestrial systems. B1.1 Background: why assess the status and distribution of biodiversity? Even given the knowledge that wetlands and their associated species are a highly valuable resource undergoing a serious ‘Biodiversity’ refers to the diversity of species of plants and decline globally, the ecological requirements for their maintenance animals on Earth. The term ‘biodiversity’, which did not come into and continued productivity are seldom included in decision- common usage until the late 1980s (Wilson 1989), includes all making processes for the development potential of wetlands. genes, species, and ecosystems, and the ecological processes For example, in China and India, where approximately 55% of which they are a part (Gaston 1996). Species are often taken of the world’s large dams are situated (WCD 2000), limited as the unit upon which assessments of the status of biodiversity consideration has been given to the downstream allocation of

32 CHAPTER 3 THE TOOLS SECTION II Jens Kipping

Chlorocypha cancellata, a rainforest species from central Africa

water for biodiversity (Tharme 2003). A high priority is now placed methods needed to collect, store, and display this information. on the development of wetland systems worldwide for provision The methods used to assess the species’ risk of extinction of water for drinking, sanitation, agriculture, and hydropower in are also described, in order to assign each species with an order to meet the Millennium Development Goals (see: www. IUCN Red List Category and determine the major threats and un.org/millenniumgoals) of improved access to water, energy and ecological requirements for each species. sanitation. With this in mind it is essential that the potential impacts of such development on wetland biodiversity, and the livelihoods Having defi ned the management issue to be addressed that depend on it, be considered within the development planning and the bounds of the study area, it is necessary to choose processes. which taxonomic groups to focus on; these should be chosen in collaboration with the livelihoods and economics team One of the major bottlenecks in bringing wetland ecosystem members, in the context of the management questions which needs into the decision-making process is a lack of readily form the focus of the study. available information on the distributions and ecological requirements of species, together with a consideration of the Ideal focus taxonomic groups might include those that are: integrating of information into decision-making processes, such ● Most easily identifi ed given the skills available as environmental fl ows (for example, see www.efl ownet.org). Even ● Most highly utilized, especially by poorer members of the where such information is made available it must be presented in communities within the project area a suitable format if the impacts of wetland development are to be ● Those where the most information already exists minimised or mitigated for. The available information on these species groups then needs In summary, the purpose of assessing the threatened status to be collated through literature review and expert opinion, such and distribution of species is to enable effective conservation of as that from local researchers and organizations, government biodiversity and livelihood values through presenting information agencies, or museum collection curators. Much information on species in a format that can be integrated into the decision- will be found in the literature; additionally some data may be making processes. The data will also serve as a baseline for available in existing databases. These sources will provide monitoring the impacts of any development or management preliminary species lists for the area, as well as information interventions, and will enable adaptive management and about the life history, habitats, and ecology of species, as evaluation of any mitigation measures put in place. well as known threats and current conservation measures. All this information can be stored within the Species Information B1.2 Overview of biodiversity methods Service database (Chapter A9), a purpose-made spreadsheet, or a GIS database. In order to demonstrate the value of freshwater species to livelihoods, we fi rst need to know what species are present, It is likely that fi eldwork will be needed to supplement the species their abundance, and where they are. This section describes the lists that have been developed through literature review and to

CHAPTER 3 33 SECTION II THE TOOLS

collect information on where (georeferenced locations, using a Defi ne study area

GPS unit) species are found. For each species group, it is likely I STAGE that a variety of sampling methods will be required. The help Use information on Choose species of taxonomic experts may be needed for species identifi cation, groups to focus on other species groups if readily available and the availability (and cost) of taxonomic experts is likely to 2 STAGE infl uence the selection of taxonomic groups that are included Literature survey – Map wetland collate species in the survey. Where species cannot be identifi ed in the fi eld, habitats data available specimens can be collected, suitably labelled, and preserved for later identifi cation. Local names can be used in place of Identify gaps in data scientifi c names, but care needs to be taken that local names refer to individual species, rather than groups of similar-looking Choose species species. The species can then be mapped to the freshwater sampling protocols and design sampling strategy habitats in which they are found using GIS. STAGE 3 STAGE

Enter data into Sample for species It is important to ensure integration between the data collected Species Information in subset of habitats by the fi eld teams. The best option is to have a fully integrated System (SIS) team i.e. a livelihoods expert accompanies the biodiversity Map species onto Conduct Red-List team and vice versa. This might not always be practicable or habitats assessment of species possible; in these cases, team members need to be aware of the information requirements of the other teams. For example, biodiversity fi eldworkers should collect relevant information Habitat maps Information about Red List status Species maps species of species

on species’ use (trade, consumption, and utilization) and be OUTPUTS sure that the information is passed on to the other teams so Synthesis with data Livelihood and that the economic and livelihood value of the species can be from other Economic maps researched and incorporated into the report. assessments

The species data collected can be used to assess the risk 4 STAGE Integration into decision-making of extinction to the species, using the Red List assessment process methodology (Chapter B11). The species information, maps and Red List status can then be combined with information from other parts of the assessment, using linking information Figure 13: Suggested planning fl ow diagram for the biodiversity such as the local names for species and the habitat areas component of an integrated wetland assessment. At each stage from which they are harvested. Following analysis, it can be it is vital that opportunities for collecting linking data to the presented in a suitable format for decision-makers, including economics and livelihoods components of the assessment are maps which integrate the information in a visually accessible followed up, and that communication is maintained with the and easily understandable way (see Chapter 6). economics and livelihoods team members

B1.3 Key resources survey can be planned. The sampling protocols required for fi sh, CBD. 2006. Guidelines for the rapid ecological assessment of molluscs, dragonfl ies and damselfl ies, amphibians, reptiles, biodiversity in inland water, coastal and marine areas. CBD birds, mammals, and aquatic plants are detailed in Chapters Technical Series No. 22/Ramsar Technical Report No. 1. B4 to B8, and general notes on species surveys are given in Joint publication of the Secretariat of the Convention on Chapter B3. If other taxonomic groups need to be surveyed, Biological Diversity, Montreal, Canada, and the Secretariat relevant protocols can often be found on the internet, in relevant of the Ramsar Convention, Gland, Switzerland. Available literature (e.g. Sutherland 2000) or by contacting experts on at: www.cbd.int/doc/publications/cbd-ts-22.pdf those species (contact the IUCN Species Survival Commission Sutherland, W.J. (ed.) 1996. Ecological Census Techniques: A to locate experts from the relevant Specialist Groups). Handbook. Cambridge University Press, UK. B2.1 Fieldwork planning B2 Planning a fi eld survey 1. Determine how much time is available for biodiversity Once the species groups to be included in the survey have surveys (i.e. number of days in the fi eld and number of been chosen and the boundaries have been defi ned, the fi eld people with biodiversity expertise)

34 CHAPTER 3 THE TOOLS SECTION II

2. Decide what other activities are needed in addition to the 14 biodiversity survey, such as documenting conservation 12

issues and threats to biodiversity, market or focus group 10 (e.g. fi sher focus group) surveys, mapping habitats (see 8 Chapters M1 to M7), collecting linking information such as 6 the local names of habitats and species etc. 4 3. Choose appropriate biodiversity survey methods and make an estimate of how long they will take found species of Number 2 4. Given the time needed to survey each site (and to travel 0 between sites), how many sites can be surveyed? Choose 0 10 20 30 40 50 60 Time (minutes) survey sites such that all wetland habitat types present are surveyed (see Chapters M5 and M6), and ensure that the Figure 14: An example of a species discovery curve, illustrating work is coordinated within the survey teams so that species the decline in the rate of accumulation of new species over time of livelihood and economic importance are included in the biodiversity survey 5. Determine whether the survey needs to be undertaken The methodology chosen will depend on the nature of the at different times of the year to take account of seasonal area and on the time, funds, expertise, human resources, and variability, such as species migrations, rainfall, and water equipment available for sampling. fl ow. For example, in low-rainfall areas, some species may be dormant and hidden for much of the year There are a wide range of existing freshwater sampling 6. In conjunction with the other members of the assessment methodologies available via the internet and a brief search team draw up a timetable of work. Ensure that time is made will produce a number of suitable methods for the taxa you for team meetings to share information and discuss issues wish to survey. Many of these will be appropriate to different that arise situations and levels of skills, funding and resources. Ensuring that an integrated approach to data collection, management, B2.2 Planning, integration and analysis and presentation is maintained is the key factor.

Figure 13 shows a suggested fl ow diagram for biodiversity B3.2 Sampling intensity and duration assessment activities. Clearly these steps do not take place in isolation, and it is important that biodiversity survey For some groups such as birds or dragonfl ies and damselfl ies, planning and activities are closely coordinated with those of timed searches may be an appropriate survey approach. Ideally the livelihoods and economics assessments, both in terms of the time given to survey should be chosen by sampling a small the data (used, for example, to inform the threat assessment number of sites intensively and recording how many species arising from unsustainable levels of utilization) and the are located per unit of time. The number of species found mapping stage, where it will be important to bring in data on over accumulated time can be plotted as species discovery the locations of area/habitats that are valuable to livelihoods. curve (see Figure 14). In this example, after 10 minutes 75% Prepare data record sheets in advance and test them during of species have been located, so you might choose to sample the pilot assessment. for 10 minutes at each location, or for 20 minutes to fi nd more than 90% of species present. The decision will depend on the For more information on the process of undertaking an available time and the extent of the area to be sampled. This integrated assessment, see Chapter 2. information can be used to decide the best use of available time to obtain suffi cient data for the maximum number of B3 Conducting species surveys sites.

This section describes general protocols for fi eld surveying. To calculate abundance, it is important that the same amount Subsequent sections describe fi eld survey methods specifi c of sampling effort (in this case time given to searching) is to the different species groups (Chapters B4 - B8). applied at each location.

B3.1 Choosing sampling protocols B3.3 What to record

Develop a standard sampling protocol for each species group The following information should be recorded for each species to be followed at each sampling location. Below are some found: suggested sampling methods for some key freshwater taxa. ● the name of the recorder and date of sampling

CHAPTER 3 35 SECTION II THE TOOLS

Suitable for use if there are BIODIVERSITY DATA COLLECTION SHEET Sheet no. few species at each location

Name of recorder Date Wetland Habitat Type

Taxonomic group(s) being sampled Sampling methods used and time/effort put in:

Records Species name Local name(s) for ID/ Location Species Specimen Photo(s) Habitat where found Notes on use, value, GPS Lat/Long OR Specimen no. species, habitat, no. identifi ed? collected? taken? and notes on ecology any other information /Way Point no. AND/OR no.s location

Figure 15: Example of biodiversity data collection sheet

● the local name of the sampling location and the species (if species) need not be collected. Species requiring identifi cation local guides are present to give this information), as well as should either be collected (storage protocols are described for additional information on the use, value and cultural role of each species group in the following sections) or photographed. the species ● the georeferenced location (determined with a GPS) B4 Fish survey sampling methods ● if the species cannot be identifi ed a specimen should be collected or a photo taken. Record the reference number of Fish are relatively easily surveyed for and are vital to nutrition the specimen or photo, the habitat the species was found and livelihoods across many parts of the world. Fish form the in, and any other useful notes on the ecology of the species, most important wetland product on a global scale providing the such as its abundance primary source of protein for nearly one billion people worldwide ● the sampling method used and the effort/time spent (FAO 2002). sampling A range of survey techniques will need to be used to obtain An example recording sheet is shown in Figure 15 (for a full a complete inventory of the fi sh species present in the survey version see Figure 49 in the appendix). This will need to be area. Local fi shermen and women can be employed to conduct tailored to meet the needs of individual surveys. the initial survey, for instance by recording their catches, or by collecting examples (voucher specimens) of the species caught B3.4 Identifi cation to species level by the fi shing community. This is an effi cient way of making an inventory of local fi sh species. Gaps in the area surveyed can It is likely that the ability to identify species will determine which be fi lled later using additional methods (for example capturing taxonomic groups are selected for survey as there is little point in less commercial species) and in additional locations, possibly collecting specimens or photographs if they cannot be identifi ed. fi shing at times not normally fi shed by local fi shers (e.g. at If good keys to species groups are available it may be possible night). Fish need be collected only if immediate identifi cation to identify species in the fi eld or later on from specimens or is not possible and specimens can be stored in either alcohol photos. Taxonomic experts can also be contacted for help, but or formalin. The methods below have been largely drawn from this should be done early, and may need to be factored into the Backiel and Welcomme (1980). budget as expertise can be expensive. Alternatively it may be acceptable to use lower levels of taxonomic identifi cation (e.g. B4.1 Market surveys family or genus) or to classify specimens into ‘morphospecies’ (see Chapter B12). Visiting markets in the area provides a good opportunity to collect integrated data, such as which species are being traded, Species which can be identifi ed in the fi eld to scientifi c name or where the species come from and who the fi shers are, and the local name (as long as the local name is specifi c to an individual value of different fi sh species (see Chapter B9). Photographs

36 CHAPTER 3 THE TOOLS SECTION II

can be taken to make a library to show when asking local or bottom-dwelling species, and a wide range of mesh sizes are people for information on when and where species are caught, needed to ensure capture of the full range of fi sh sizes present. and how much they are bought and sold for. They are suitable for collecting qualitative information on the species present, as required during rapid species assessments, Following these initial surveys researchers can accompany and can easily be placed in a wide range of freshwater habitats. fi shers to fi shing sites to sample their catches and to collect location data (using a GPS) on where species are caught. Gillnets vary widely both in their physical structure (dimensions, colour, mesh size, twine material and thickness, hanging and B4.2 Conducting a fi sh survey rigging of weights and fl oats) and in how they are set (perpendicular or parallel to shore; in straight lines, zig-zags or looped to form Local fi shers may not fi sh in all the wetland habitat types present traps; anchored in place or drifted with currents). The choice of net within a survey area, so some habitats may have to be sampled types and method will depend on the type of water and species of separately. Either local fi shers can be employed to collect fi sh fi sh to be sampled. samples in these areas using their own gear under the guidance of the survey leader, or separate fi sh surveys can be conducted. Seine nets are suitable for collecting rapid samples but can be used only where the river or lake shore grades into a hard, gently sloping A variety of fi sh survey methodologies are summarised below. bottom with no obstacles such as rocks or submerged branches. The choice of method and how it is employed will depend on When skilfully employed they can capture the majority of fi sh within the habitat being sampled, and water depth, clarity, fl ow, and the sample area. However they are expensive unless they can be vegetation will need to be considered (Côté and Perrow 2006). rented from local fi shermen, and a boat is usually required to take Fishing equipment used by local fi shers can be used for the the net out in a sweep of the area being sampled. survey, but it is desirable to use a range of sampling methods to overcome method-specifi c biases, to conduct day and night Cast nets can be employed to fi sh in most wetland habitats but sampling, and to sample in places where less commercial species they require a certain degree of skill for effective use. are found. B4.2.2 Other methods B4.2.1 Nets Traps come in a wide range of sizes and designs including small Gillnets are versatile, low cost, and easy to operate. They can be ‘basket traps’ and ‘fence traps’ which direct the fi sh into baskets. used in lakes of any size, in deep or shallow water, over bottoms Local fi shermen will often have designed traps most suitable for too rough for seine nets, and on a large or small scale. Their the area to be surveyed. main disadvantage is that they may not catch largely sedentary A hook and line is one of the most common methods used for catching fi sh. Requiring only a single baited hook and line, it is cheap and easy to use. Alternatively long lines of hooks can be used, and these may be left tethered to posts for a period of time or overnight. This method is selective for carnivorous species that readily take the bait.

Electrofi shing requires specialised equipment operated by trained personnel. It is quick, requires few people and little physical exertion; however it is dangerous for both fi sh and operators, and the equipment is expensive. It is mainly suitable for use in fl owing water less than 2 m deep.

Kong Kim Sreng/Darwin Integrated Wetland Assessment Project Kong Kim Sreng/Darwin Integrated Wetland Explosives and poisons such as rotenone should not be used.

B4.3 Where to sample and how to standardize fi shing effort

The full range of wetland habitats present should be sampled, as described in Chapter M5. Within each habitat type, it is recommended to sample from as many sub-habitats as possible to get comprehensive species lists (within a lake for example, there Fish traps in Stung Treng Ramsar Site, Cambodia may be shallow vegetated areas, deep areas and rocky shores).

CHAPTER 3 37 SECTION II THE TOOLS Kong Kim Sreng/Darwin Integrated Wetland Assessment Project Kong Kim Sreng/Darwin Integrated Wetland

Gastropod collection by villagers in Stung Treng Ramsar Site, Cambodia. Molluscs were also observed on sale in the local markets, and the project collected data on prices both within villages and at local markets

Sampling effort can be standardized using Catch Per Unit Effort Côté, I.M. and Perrow, M.R. 2006. Fish. In: Ecological Census (i.e. how much is caught by fi shing for one man-day in each habitat Techniques: A Handbook (ed. W.J. Sutherland); 2nd edition. using the same fi shing techniques). Cambridge University Press, Cambridge, UK. Nielsen, L.A. and Johnson, D.L. (eds.). 1983. Fisheries Techniques. B4.4 Preparing specimens and identifi cation American Fisheries Society, Bethesda, USA. Sutherland, W.J. 2000. The Conservation Handbook: Research, Where fi sh can be identifi ed to species on-site there is no need Management and Policy. Blackwell Publishing, Oxford, UK. to collect specimens. If there is uncertainty as to the identifi cation of the fi sh, a mature adult specimen of each species should be B5 Mollusc sampling methods collected if possible. Fish should be killed using an anaesthetic such as benzonocaine if this is available, then preserved in either B5.1 Introduction formalin or alcohol. Formalin is simple and cheap, but toxic to humans, so alcohol may be preferred. Fix fi sh in 70% alcohol Freshwater molluscs provide vital additional nutrition in many parts before storing them in 40% alcohol (Sutherland 2000). Attaching of the world, and play a key role in maintaining wetland ecosystems a permanent label directly to the specimen, or placing fi sh and through their control of water quality and nutrient loads. They are labels in pierced plastic bags within a larger container of formalin or however one of the most threatened groups of freshwater taxa, alcohol avoids the need for several individually labelled containers. with key threats including pollution, dams, drainage, and siltation. The colours of fi sh should be noted or photographed, as they are removed by alcohol. For large fi sh which cannot be collected for The degree of utilization of molluscs by people for food (or, practical reasons, photos should be taken, including diagnostic occasionally, for fi shing bait and other purposes) varies greatly features and an object for scale (such as a ruler). across the world. Where molluscs are consumed a market survey (Chapter B9) may reveal if they are traded, and which species are B4.5 Key resources utilized and are preferred.

Backiel, T. and Welcomme, R.L. 1980. Guidelines for sampling fi sh Freshwater molluscs are typically divided into two groups; in inland waters. EIFAC Technical Papers (EIFAC/T33). Available gastropods (typical snails) and bivalves (mussels for example). The at: www.fao.org/docrep/003/AA044E/AA044E00.htm following sections give methods for surveying each of these.

38 CHAPTER 3 THE TOOLS SECTION II

B5.2 Gastropods B5.2.4 Preparation of specimens for storage and identifi cation Specimens should be cleaned after collection to remove as much Gastropods can be collected using quadrats, sweep netting debris and as many other organisms as possible. Specimens are through vegetation, dragging a hand-net over the under-water more easily observed and sorted if they are submerged in clean substrate surface and washing/scrubbing rocks. The various water. Relaxation (immersing the specimen in a water/menthol methods described are suited to different environments. solution overnight) is used to encourage the snail body to come out of the shell, making the soft parts available for species B5.2.1 Quadrat sampling identifi cation. Quadrat sampling is suitable for shallow, slow-fl owing areas, as well as for shallow edges of pools and lakes. A quadrat can be Where samples contain large volumes of substrate (sand or fi ne made locally (a simple square frame of a standard size, often 0.5– gravel) and small numbers of molluscs, separation and relaxation 1.0 m) and constructed from any rigid material, sometimes with of specimens is not practical. Preserve the sample in the fi eld a grid of wire or string to subdivide the quadrat into small, more (using either 70% isopropyl or ethyl alcohol). The sample should easily searched squares. then be re-sieved in the laboratory to remove fi ne sediment and plant and animal detritus, and the sample examined through a low- A series of quadrat samples ranging from a minimum of eight to power binocular microscope for small or inconspicuous molluscs. as many as 16 should be collected from within each sampling site to produce a total area sampled equal to about 5–10 m2 for For long-term preservation, the specimens should be placed in each habitat location. Quadrat samples may be: i) concentrated a solution of 70% ethyl alcohol, 15% glycerin, 15% water, and in areas perceived as representing the most suitable habitat to buffered to pH7. While it is preferable to keep the soft parts of enhance the possibility of detecting the target species; or ii) placed snails, if it is considered suffi cient to identify gastropods only to systematically along a river or lake shore if the area appears to be genus or family (e.g. in a rapid assessment) just the shells can be relatively homogeneous. kept. To remove the soft parts, place the snails in boiling water and then pull the soft parts out of the shell with forceps. In coarse substrate areas, molluscs should be either hand collected or brushed from individual stones into a tray, net or sieve. B5.3 Large freshwater bivalves (more than 2.5 cm in The bedrock or stones can be scrubbed underwater with a brush length) so that dislodged snails are swept into a submerged net or sieve placed downstream. Alternatively, rocky substrates can be placed Larger bivalves tend to be found in shallower areas, although they in a tray underwater and carried to a more convenient location for may also be found at lower densities at greater depths. If a boat is processing. available, dredging is probably the quickest and easiest method. Otherwise hand-sampling or using a hand-net from the bank are Areas with fi ne substrate (such as muds, sands, or silts) are sampled the best methods. by excavating bottom sediment from within the quadrat to a depth of about 3 cm using a dip net or sieve with an effective mesh size B5.3.1 Dredging of 0.5 mm or smaller. The sample should be washed through a A dredge (Figure 16) can be used to collect large freshwater bivalves sieve to remove as much substrate as possible. Generally a 0.25– either by throwing it into the water from the bank or by pulling it 0.5 litre volume of sieved ‘concentrate’ from each such site is an along behind a boat travelling upstream. The mesh size defi nes the adequate sample. size of the smallest bivalve collected, and its use may be limited by the substrate, depth, and fl ow of a river. To standardize sampling, B5.2.2 Sweep-netting it is recommended to drag the dredge across a standard distance Areas with rooted aquatic vegetation may contain large numbers a fi xed number of times (fi ve transects of 10 m for example) at of gastropods. In shallow areas a hand net can be swept through specifi ed points in the river. Alternatively a fi xed sampling time the vegetation, and the vegetation vigorously shaken to dislodge can be used. It is not a very quantitative sampling method but is molluscs. In deeper waters a grapnel (a weighted three-way hook generally quick and easy in shallow waters (less than 8 m) where on a rope) will bring vegetation to the surface, which can then be most mussels are found. washed into a bucket to retrieve attached gastropods. The number of sweeps should be standardized between sites such that the B5.3.2 Using a grab sampling effort is equal. Grabs are more quantitative than dredges but sample a smaller area of substrate, so more grabs are required to sample the B5.2.3 Other methods substrate suffi ciently to detect most of the species present. They Some gastropods will also be found using the methods for work at greater depths and higher fl ows than dredges. They are less sampling small bivalves, as described below. effective on some substrate types such as very fi rm substrates. A

CHAPTER 3 39 SECTION II THE TOOLS

mussels can be placed in boiling water until they open and the soft parts removed. Recently-dead mussels are often found, so it may not be necessary to kill live specimens. Both valves (shells) should be kept and held together to enable identifi cation.

B5.4 Smaller freshwater bivalves (less than 2.5 cm in length)

Smaller bivalves can be collected by a wide range of sampling methods including netting, sweeping submerged vegetation (as Figure 16: A hand-dredge for mollusc sampling described for gastropods), or kick-sampling (see B6.4.1).

B5.4.1 Hand-netting standard number of grabs should be taken from each sampling An ideal hand-net to use for this purpose is a robust, aluminium- point, and the area sampled can be calculated from the area of framed pond net (approximately 0.4 m square), with a nylon mesh gape of the grab. Grabs are often heavy and unwieldy, so need bag (0.3 m deep, 0.5 mm mesh; Figure 17). Most bivalves live close to be used from a relatively sturdy boat. A winch or pulley system to the surface of the substrate and can be collected by skimming may be needed for retrieval as the grab needs to be heavy to the sample net through the top 2–3 cm of sediment from the bank ensure penetration of the substrate. or a small boat. Agitate the net in the water to sieve out mud and silt. The material can then be washed into a white sorting tray or B5.3.3 Hand-sampling bucket before passing it through a 4 mm sieve to collect the larger This is only feasible in the shallow margins of rivers where it is specimens and to remove coarse debris, and then through a 0.5 possible to easily reach the bottom substrate. However, these mm sieve to collect remaining bivalves. Specimens can be picked areas often contain the highest densities of mussels. Hand from the sediment by examining a small quantity in a glass dish sampling can be made quantitative by either sampling within under a binocular microscope at x6 to x10 magnifi cation. quadrats or doing timed searches. It is suitable in both very turbid rivers with muddy substrates and clear waters, where mussels B5.4.2 Dredging may be located by sight (e.g. using a glass-bottomed bucket). For water bodies deeper than 1.5 m, samples can be collected Where mussels are at relatively high densities it is the quickest using a hand dredge (as described in B5.3). Although these are and easiest method of sampling. usually equipped with a relatively coarse-sized mesh (>4 mm), on soft substrates they rapidly become clogged with fi ne sediment B5.3.4 Using a hand-net so can be used only over short distances. The captured sediment If the water is shallow and easily accessible, mussels can be can then be passed through sieves in order to pick out smaller sampled using a standard hand-net with a relatively large mesh bivalves. bag, which is dragged across the substrate surface either from the bank or from within the water. However if mussels are present B5.4.3 Processing and storage of bivalves at low densities they may not be detected using this method. This Samples can be stored in water or preserved in alcohol. Small method can be made quantitative by doing timed searches. bivalves will remain fresh for 3–4 days when stored in their

B5.3.5 Scuba-diving This is expensive and often not practical, requiring a lot of expertise, expensive equipment, and presenting various safety issues. It is widely used in North America in relatively shallow rivers with very low turbidity so that mussels can be searched for by sight using timed searches.

B5.3.6 Preserving mussels for later identifi cation Alana Ecology www.alanaecology.com Mussels should be rinsed with water to remove mud. Mussel identifi cation is often by shell characters, so the soft parts may not be needed (check identifi cation keys if available for local species). If the soft parts are required, preserve in 95% ethanol (which should be changed after a couple of days as the mussels may release lot of water on opening). If only the shells are required, live Figure 17: A hand-net for sampling small bivalves

40 CHAPTER 3 THE TOOLS SECTION II

native water and kept in a refrigerator. If live specimens are to be returned to their original habitat they should be examined under a cold light source – a short period out of water will not kill them. For Jens Kipping longer term storage they should be preserved in 70–80% alcohol (Industrial Methylated Spirit, IMS). Alternatively specimens may be placed on absorbent paper and allowed to air-dry.

Identifi cation sometimes requires internal examination of the hinge features which requires the separation of the two valves. For freshly collected specimens and those preserved in alcohol, the valves may be opened and the animal removed by immersing in boiling water. Articulated specimens that are totally dry can also be boiled to separate the valves but some may require chemical treatment by placing them in a solution of domestic bleach (50/50 with water). This will dissolve the hinge ligament but the periostracum (a thin organic coating or ‘skin’ which is the outermost layer of the shell) and soft parts of the animal will be destroyed. As soon as the Dragonfl y and damselfl y sampling in the Okavango Delta treatment is complete the separated valves should be washed in water to remove the bleach and allowed to dry. The bleaching will whiten the shell and enable features of the hinge line to become their lifecycle. Many species spend part of their lifecycle within more clearly defi ned. woodlands for example, or forage within non-wetland habitats. It is important to include a range of habitat types if a comprehensive B5.5 Further information survey is being attempted.

A comprehensive guide to sampling for freshwater mussels B6.2 Sampling for adults is given by Strayer and Smith (2003). Information on collection of smaller bivalves has been taken from a protocol used for B6.2.1 Habitats sampling small bivalves in the UK (Killeen et al. 2003) which may Dragonfl ies and damselfl ies occur in all types of freshwaters and need alterations for use in larger tropical rivers. The information on in nearby habitats. They often prefer sunny places where they can gastropod sampling has been taken from Furnish, Monthey and bask, but there are also species which live in shade. Wide-ranging Applegarth (1997). species may also be found in temporary or disturbed habitats such as puddles, rice fi elds and ditches; specialist and endemic species B5.6 Key resources are more likely to be found in pristine forest and wetland habitats and in small micro-habitats such as seepages (where water oozes Furnish, J., Monthey, R., and Applegarth, J. 1997. Survey Protocol from the ground), the spray-zone of waterfalls, wet trickles on rock for aquatic mollusk species from the northwestern forest plane. faces, torrents, small pockets of water in tree holes (phytotelmata) Version 2.0 - October 29, 1997. U.S Department of the Interior, or small pools and swamps in forest. As many potential habitats Bureau of Land Management. Accessed on 22/1/2009 at www. should be sampled as possible, not only river banks. blm.gov/or/plans/surveyandmanage/SP/Mollusks/acover.htm Killeen, I., Aldridge, D., and Oliver, G. 2003. Freshwater Bivalves B6.2.2 Survey methods of Britain and Ireland. Occasional Paper 82. Field Studies Dragonfl ies and damselfl ies may be surveyed by collecting or by Council, UK. observation using close-focus binoculars; however some species Strayer, D. L. and Smith, D. R. 2003. A guide to sampling freshwater (especially in the tropics) are diffi cult to differentiate visually and mussel populations. Am. Fish. Soc. Mono. 8:1-103. the collection of voucher specimens is recommended. To catch them use a large hooped net on a long stick (a 40–75 cm diameter B6 Dragonfl y and damselfl y sampling methods hoop with a handle 1–2 m long is suitable with extendable poles if possible). The netting is usually white, green or black and the bag B6.1 Introduction of the net needs to be deep enough to fold it closed, so that the dragonfl y is not able to escape when you fl ip the rim over the net Dragonfl ies and damselfl ies (Odonata) are relatively easy to sample, to trap it in the bag. requiring limited equipment, and can be a valuable indicator of overall ecosystem health. Adult dragonfl ies are not restricted The most effective technique is to wait until the adult dragonfl y is to wetlands, but all larvae are aquatic and water is essential to just past you, and then swing the net from behind. Some species

CHAPTER 3 41 SECTION II THE TOOLS

is suitable, not that sold for cosmetic use) for 12-24 hours. Remove from acetone and dry in air. If pure acetone is not available, the specimens can be killed in alcohol and then dried well with silica gel or in the air in arid environments.

B6.3 Sampling for exuviae

Exuviae are the casts skins of the penultimate larval stage of dragonfl ies and damselfl ies; some (very few tropical species) can be identifi ed to species level. They can also provide useful information about where species breed. Good places

David Allen/Darwin Integrated Wetland Assessment Project David Allen/Darwin Integrated Wetland to look for exuviae include rocks along the edge of the water, debris sticking out of the water, emergent aquatic vegetation such as reeds and rushes, tree snags and branches, wooden posts, bridge abutments, pilings and so on. Generally exuviae are found only a few inches above the level of the water, but occasionally they may be up to 2 m above the water level. They are easiest to fi nd when looking from the water towards the shore. No special equipment is needed to collect them but Dragonfl ies can be held between the fi ngers or thumb and fi nger for ensure that they are well dried before storage. examination and photographing, and then released unharmed. This is an adult female Anax tristis from Dai Lake, Mtanza-Msona, the fi rst B6.4 Larvae record for this species from the Rufi ji District, Tanzania Damselfl y and dragonfl y larvae are aquatic and are most commonly found in ponds, marshes, lake margins, shallow are more easily caught when they alight on a perch or while areas of streams and the slower reaches of rivers and streams, basking on logs, or at certain times of day. Watching the habits or in water-fi lled hollows within trees. Some species occur in of a species before trying to catch it will yield greater success. brackish pools and estuarine habitats. Larvae are most easily When possible, only sample mature males to minimise impacts collected by kick-sampling (below) in shallow areas or sweep- on breeding populations. netting amongst aquatic vegetation. Some may also be caught by dredging (for example when surveying for bivalves). To reduce the number of specimens collected, develop a reference collection of species present. Once familiar with local B6.4.1 Substrate sampling species it may be possible to record species by observation only, Small pools are best sampled with a small dip-net (or a kitchen or by catching them and re-releasing them. Even if a species’ sieve can be used), while rivers are best sampled with a hand- scientifi c name is not known the species may be recorded by net or kick-seine. Kick-sampling involves placing a net about referring to a reference specimen which will later be identifi ed 30 cm downstream and disturbing the substrate with the feet. to species.

Once caught dragonfl ies are best held with the wings folded together between the thumb and forefi nger (or two fi ngers). Larger species can be held at the thorax or legs provided at least three legs on one side are grasped. If handled carefully most individuals will fl y off unharmed once released (Dijkstra 2006).

B6.2.3 Preservation Make a note of the specimen’s colours (particularly eye colour) or take a photograph (as the colours can fade on storage) and place the specimen briefl y in acetone to kill it and then place the wings together and straighten the abdomen. Place the specimen in porous paper envelopes (Figure 18), recording the specimen number, date, photograph details, and collection location on the envelope. Place in acetone (only pure acetone Figure 18: Template for making paper triangles

42 CHAPTER 3 THE TOOLS SECTION II

discussions on the best nets to use and preservation techniques. Notes on kick-sampling can be found at www.environment.fi and in the Western River Basin District Project (Walsh 2005).

B6.6 Key resources

Bright, E. 1999. Sampling Protocol for Odonata Larvae. Michigan Odonata Survey Technical Note No. 2. Insect Division, Museum of Zoology, University of Michigan, U.S. Available at: http://insects.ummz.lsa.umich.edu/MICHODO/ mospubs/MOSTN2.pdf

David Allen/Darwin Integrated Wetland Assessment Project David Allen/Darwin Integrated Wetland Dijkstra, K.-D. B. and Lewington, R. (illus.). 2006. Field Guide to the Dragonfl ies of Britain and Europe. British Wildlife Publishing. Gillingham, Dorset, UK. Walsh, A. 2005. Small Streams Risk Score Method Manual. Western River Basin District Project, Galway County Council, Ireland. Available at: www.wrbd.ie/PDF/SSRS-Training- manual_11_01_06.pdf Air-drying adult dragonfl ies after immersion in acetone for 24 hours B7 Sampling methods for non-fi sh vertebrates associated with wetlands: Organisms that are dislodged will be collected by the net or amphibians, birds and mammals screen as they are washed downstream. Empty the net into a pan or screen to pick out larvae. The hand-net can also be used B7.1 Introduction to sample underneath undercut banks and to sweep through aquatic vegetation growing in slow-moving or still portions of Non-fi sh vertebrates such as amphibians, reptiles, birds, and the stream or river. Sample among and underneath woody or mammals can be used as indicators of the ecological integrity leafy debris accumulations as these habitats often harbour a of wetland habitats. They can be used to prioritize wetland great number of specimens (Bright 1999). habitats for conservation, and the relative importance of different sites (for breeding, feeding, or resting) can be determined. In Preserve specimens in 70% alcohol. Do not put too many many instances local communities depend on these groups specimens in a container as they may damage each other as supplementary food resources, and for income through before they die. If a lot of debris is present in the container with the trade in bush meat, traditional medicines, the pet trade, or the organism consider using 95% alcohol to compensate for other animal parts such as fur and skins. As with other taxonomic dilution. In either case, replace with fresh alcohol frequently. groups market surveys and community questionnaires can be invaluable for collecting information on traded species as well as B6.5 Further information quantities and values.

There is extensive information on the internet describing how A variety of sampling techniques can be used to document species to sample for dragonfl ies and damselfl ies. The Asia Dragonfl y composition, richness, density and relative abundance of non-fi sh website (www.asia-dragonfl y.net) provides an excellent guide vertebrates associated with wetlands. Field guides or identifi cation by Viola Clausnitzer, KD Dijkstra and Vincent Kalkman (follow keys will facilitate the identifi cation of individual species in the fi eld the link labelled How to: Studying Tropical Dragonfl ies and and are available for most countries or regions for both birds and Damselfl ies). The Michigan Odonata Survey (http://insects. mammals, and to a lesser extent for amphibians and reptiles. If ummz.lsa.umich.edu/MICHODO/mospubs/) has several useful a species cannot be identifi ed in the fi eld a specimen collection technical notes, such as Collecting Specimens for the Michigan might be considered, but the impact of collection on the local Odonata Survey; Odonata Collecting Instructions; Sampling population should be assessed. Protocol for Juvenile Odonata; and Preserving Adult Odonata. B7.2 Amphibian and reptile survey methods The International Odonata Research Institute’s Odonata Information Network (www.iodonata.net) has several useful Many species of amphibians and reptiles (herpetofauna) tend to pages particularly the Collecting and Preserving Dragonfl ies be nocturnal so night sampling will be required. Amphibians in Frequently Asked Questions page which has extensive particular may be low in abundance during drier seasons, and

CHAPTER 3 43 SECTION II THE TOOLS

Table 5: Standard sampling techniques to record herpetofauna

TECHNIQUE INFORMATION GAINED TIME x COST y PERSONNEL z

Visual encounter surveys Species richness Low Low Low Quadrat sampling Density, relative abundance and species richness High Low Medium Transect sampling Density, relative abundance and species richness High Low Medium Drift fences and pitfall traps Relative abundance and species richness High High High x Relative time investment y Relative fi nancial costs: High – expensive; Medium – moderately expensive; Low – relatively inexpensive z Personnel requirements: High – more than one person required; Medium – one or more persons recommended; Low – can be done by one person

sampling should preferably be undertaken during the wetter B7.3 Bird survey methods seasons, which usually coincide with their breeding period. Some standard sampling techniques to record amphibians and reptiles Birds, being generally conspicuous, are relatively easily surveyed are highlighted in Table 5 (adapted from Heyer et al. 1994). and counted, though some skill in identifi cation, especially of bird calls, is required. Some standard sampling techniques to record A visual encounter survey (VES) is the easiest and lowest cost birds, and their costs and benefi ts, are highlighted in Table 6 technique to document amphibians and reptiles associated with a (adapted from Sutherland 2000, and Sutherland et al. 2004). wetland. This involves one or more people walking through an area or habitat for a prescribed time period, systematically searching Many bird species are highly seasonal, either moving between for amphibians and reptiles. Time is expressed as the number of seasons to follow suitable habitat, or as part of larger migratory person-hours of searching in each area. A VES can be easily carried movements. This is especially true in the case of many wetland out in a number of quadrats along a transect of specifi ed distance. birds and it is important that this is taken into account when designing the survey timetable. In general, sampling should be Quadrat sampling consists of laying out a series of small quadrats undertaken during both drier and wetter times of the year. (or strip quadrats) at randomly selected sites within a habitat and thoroughly searching each quadrat for amphibians and reptiles. The species discovery curves (SDC; see Figure 14) and McKinnon The quadrats should be separated by adequate distance to avoid Lists (ML; McKinnon and Phillips 1993) involve similar techniques presampling disturbances. Quadrats can vary in size between 1 where the cumulative (total) number of species recorded is plotted x 1 m and 8 x 8 m according to the density of amphibians and against sampling effort (i.e. number of observer hours/days for reptiles in a particular locality – use a larger quadrat if the animal SDC, and number of lists of 20 bird species for ML). The McKinnon density is low. Lists method enables the comparison of bird species richness in different sites through the curves in the plot. The species discovery Amphibians and reptiles tend to respond differentially to curve for a particular site shows the point at which further effort is environmental gradients governed by moisture, vegetation cover unlikely to reveal further species in a particular locality. and so on. The transect methodology can be used to sample either across these habitat gradients or within habitat types, where randomly located narrow linear strip transects (i.e. 2 x 50 m, or 2 x 100 m) are laid out, and the portions of habitats within the transect are thoroughly searched for herpetofauna.

The drift fence and pitfall trap method involves the use of drift fences (low barrier made from plastic or fabric 0.4–0.5 m in height and 5–50 m in length) that direct animals into traps placed on either side of the barriers. The traps can be pitfalls (made from buckets or plastic pipes, for example), funnel traps or a combination of the two. Drift fences and pitfall traps can be placed around ponds, marshes, and in stream/river banks, arranged either in a linear

manner, or in a combination of arrays. Assessment Project Gita Kasthala/Darwin Integrated Wetland A drift fence with bucket traps being installed along a lake shore

44 CHAPTER 3 THE TOOLS SECTION II

Table 6: Standard sampling techniques to record birds

TECHNIQUE INFORMATION GAINED TIME x COST y PERSONNEL z

Species discovery curves Species richness Low Low Low McKinnon Lists Species richness Low Low Low Timed point counts Density, relative abundance and species richness High Low Medium Line transects Relative abundance and species richness High Low Medium x Relative time investment y Relative fi nancial costs: High – expensive; Medium – moderately expensive; Low – relatively inexpensive z Personnel requirements: High – more than one person required; Medium – one or more persons recommended; Low – can be done by one person

A point count is a count of species (and individuals) undertaken ● key informant interviews from a fi xed location for a fi xed time period (for example 10–20 ● focus group interviews (e.g. traditional medical practitioners) minutes). Points should be at least 200 m apart to prevent double counting. Line transects involve observer(s) moving along a fi xed Wetland walks can be invaluable for collecting information route and recording the birds they see on either side of the route. about which plants are utilized; visit a chosen range of wetland Transects can be carried out by walking on land or by boat. The habitats with local people and ask them which plants are used total transect length will vary according to the size of the wetland, and what for (using standard ethnobotanical techniques). Plants and individual transects range from 100–1000 m. It is also possible which they point out as being important to local livelihoods to conduct timed point counts at fi xed distances along a line can then be identifi ed (if a taxonomic expert is present), or transect or in a range of habitat types. collected for later identifi cation (as described below). Such an approach is recommended where time is limited, providing B7.4 Key resources information which is suitable for integration with the economics and livelihoods data. Heyer, W.R., Donnelly, M.A., McDiarmid, R.W., Hayek, L.C. and Foster, M.S. (eds). 1994. Measuring and Monitoring Biological If more time is available, it may be possible to do a more thorough Diversity: Standard Methods for Amphibians. Smithsonian survey of the aquatic plants of the area. The aquatic fl ora can Institution, USA. be roughly divided into macroalgae, submerged vascular plants, McKinnon, J. and Phillips, K. 1993. A fi eld guide to the birds of emergent vascular plants and bank-side vegetation, with a Borneo, Sumatra, Java and Bali. Oxford University Press, possible fi fth category of seasonally-fl ooded terrestrial plants. Oxford, UK. Sutherland, W.J. 2000. The Conservation Handbook: Research, Bank-side fl ora and seasonally-fl ooded terrestrial fl ora may be Management and Policy. Blackwell Publishing, UK. surveyed by establishing transects with a rope and identifying Sutherland, W.J., Newton, I and Green, R.E. 2004. Bird Ecology all plants to a certain distance on either side of the transect. and Conservation: A Handbook of Techniques. Oxford The transect length and width will depend on the time available University Press, UK. for the survey – a standard length is 100 m. Several shorter transects randomly spaced throughout a habitat are preferable B8 Plant survey methods to one long transect, but a long thin transect is preferable to a short broad transect. Alternatively, quadrats may be marked out B8.1 General approach at randomly selected (see Sutherland 2000) locations across a site and all plants with their roots within the quadrat recorded. Because of the high diversity of wetland plants it will probably A larger number of smaller quadrats are preferable to a small be necessary to restrict surveys to aquatic plant species of direct number of large quadrats. importance to humans, such as plants used for food, animal fodder, or construction materials. To discover which wetland Similar approaches may be used for submerged and emergent plants are used researchers can adopt a mixture of approaches, vegetation, where transects may be marked out in the water using linking in with the work of the livelihoods team: buoys (these can be made from an empty bottle or a balloon ● wetland walks attached to a rock with a rope whose length is approximately the ● local market surveys same as the water depth), and all plants that are visible from a ● household interviews boat or collected with a grapnel along the transect are recorded.

CHAPTER 3 45 SECTION II THE TOOLS

Alternatively, sampling can be done from predetermined, are likely to fade on drying, especially fl owers), local names and randomly-chosen locations in the water located using a GPS, use, if any. In the fi eld, specimens should be stored in a press either from within a set area (e.g. an imaginary 3 x 3 m quadrat (see below) as soon after collecting as possible. Place the plant next to the boat) or with a standardized number of throws of fl at between sheets of newspaper with layers of corrugated the grapnel (Madsen 1999). It is also possible to survey along cardboard, if available, between the plants to allow air to get transects laid out perpendicular to the shoreline, thereby into the stack. The plants should be arranged in a way which encompassing all forms of aquatic vegetation. demonstrates the characteristics necessary for identifi cation (i.e. showing both sides of leaves and the underside of fl at Emergent, bank-side, and terrestrial plants can be collected by fl owers) as well as fruit and seeds. In damp areas or where hand. Macroalgae are often found in mats at the surface and it may be some days before the specimens can be properly may also be collected by hand. Submerged vegetation and dried in a herbarium, or if pressing succulent plants, the paper deeper algae may be collected using a grapnel or any kind of should be replaced every few days. Fruits may be dried whole, weighted hook or rake attached to a rope. Alternatively a dredge or sliced and pressed, or preserved in 70% alcohol and stored or grab may be used; these are likely to damage plants, but separately, ensuring that they are clearly labelled. Cones and may bring up tubers or rhizomes which could be useful in plant wood should be labelled and air-dried. As they are collected, identifi cation. Diving is also an effi cient method of surveying place the stack of specimens between two boards, kneel on submerged aquatic vegetation, although it may be costly and the entire stack and tighten with straps or a rope around the requires divers who are suffi ciently qualifi ed and experienced. press.

B8.2 Collection and storage of plants Plant presses can be purchased, but can also be easily made from rectangles (approximately 75 x 75 cm) of hardwood or Plants which cannot be identifi ed in the fi eld should be collected plywood board or a wooden lattice (good for allowing more for later identifi cation. Aim to collect healthy, full-sized leaves rapid drying of the specimen). Adequate small presses can be still on the stem, as well as any fl owers and fruiting bodies. For made from wire grids, such as a cake tray. trees, it may be helpful to collect a small specimen of bark. For each sample record the date, name of collector, and location If drying is not possible in the fi eld, stacks of plants pressed (name and GPS location, and altitude) where the sample was within newspaper can be sprayed with alcohol or a litre of taken, as well as additional information on colours (as these 70% alcohol can be poured over a 20 cm stack of specimens kept sealed in a plastic bag. The resulting specimens may be blackened and brittle and need to be checked on a regular basis to be sure that they are not heating up, but the specimens will still be satisfactory for identifi cation.

Succulents should be killed by submergence in boiling water for a few seconds as the tissue will then dry more quickly and it will also prevent them growing new shoots in the press.

Mosses are usually placed directly into a paper packet for drying and are not pressed. Liverworts tend to shrivel so some gentler pressing is sensible. Lichens, collected on their substrate if possible (for example, cut on a sliver of bark from

David Allen/Darwin Integrated Wetland Assessment Project David Allen/Darwin Integrated Wetland a tree), can be simply air dried in most cases and do not require pressing. Mosses, liverworts and lichens are usually stored in paper packets. Macroscopic algae can be pressed and dried, freeze dried or stored in 40% alcohol (although they lose their pigments in alcohol). Flimsy algae are best placed The heart-shaped leaf in the centre of the image is an aquatic plant on a herbarium sheet underwater and then gently lifted and Ipomaea aquatica collected from the margins of the Rufi ji River and drained. utilized as a vegetable by the Mtanza-Msona villagers. The plant in the background (Water lettuce Pistia stratiotes) is an introduced invasive If specimens are required for long term storage in a herbarium, aquatic plant found in small clumps and large dense mats across the rather than simply for identifi cation, then further treatment will Mtanza-Msona wetlands, both in the Rufi ji River and in lakes. Dense be required, and advice should be obtained from a herbarium mats of Pistia disrupt fi shing activities, especially in the lakes curator. Victor et al. (2004) (www.sabonet.org.za/reports/

46 CHAPTER 3 THE TOOLS SECTION II

publications_report25.htm) and Bridson and Forman (2004) are Publishing, The Hague, The Netherlands. excellent resources on the collection of plant specimens and the Madsen, J.D. 1999. Point and line intercept methods for aquatic development of a herbarium. plant management. APCRP Technical Notes Collection (TN APCRP-M1-02), U.S. Army Engineer Research and B8.3 Identifi cation Development Center, Vicksburg, USA. MEA. 2005. Ecosystems and human wellbeing: Wetlands and Depending on the skills of the assessment team members, plants water Synthesis. World Resources Institute, Washington, can either be identifi ed in the fi eld with fi eld guides and keys DC. Available at: www.millenniumassessment.org/ where they exist (e.g. Cook 1996), or later with the assistance of documents/document.358.aspx.pdf herbarium staff. Ramsar Handbooks for Wise Use. Available at: the Ramsar Wise Use Resource Centre www.ramsar.org/wurc/wurc_ Field guides are available for some families of plants and for index.htm some geographical regions. Plant identifi cation keys are available Sutherland, W.J. 2000. The Conservation Handbook: Research, for many more plant families, but often require a higher level of Management and Policy. Blackwell Science, UK. botanical knowledge to use. Increasingly, plant keys, fl oras and Victor, J.E., Koekemoer, M., Fish, L., Smithies, S.J. & Mössmer, other resources are becoming available on the internet, often M. 2004. Herbarium essentials: the southern African produced by the major botanical gardens. For example: herbarium user guide. Southern Africa Botanical Diversity Network Report (SABONET) Report No. 25. National ● Interactive key to the rattans of Lao PDR Botanical Institute, Pretoria, South Africa. Available at: www.kew.org/data/rattanslao www.sabonet.org.za/reports/publications_report25.htm ● Interactive key to the fl owering plants of the Neotropics www.kew.org/science/tropamerica/neotropikey.htm B9 Market surveys ● Aluka Africa plant resource database www.aluka.org/page/content/plants.jsp A full presentation of market survey methodologies is beyond ● eFloras the scope of this toolkit, but an awareness of some of the www.efl oras.org issues that should be considered when planning a market survey is important. B8.4 Further information B9.1 Introduction For more information on line transects and point sampling see Madsen (1999). In the context of biodiversity assessment we defi ne market survey here as meaning a survey of the physical market, For the identifi cation of aquatic plants Cook (1996) is an excellent the location where natural resources and products are sold. resource, with a key covering the vascular aquatic plants of the However, a market survey may also encompass research into world. the structure and institutions of a market from an economic and sociological perspective to understand how the market B8.5 Key resources functions, who has access to the market to buy or sell, and so on. Borrini-Feyerabend, G., Pimbert, M., Farvar, M.T., Kothari, A., and Renard, Y. 2004. Sharing Power: Learning by Doing in Surveys of markets (including informal bartering or exchange Co-management of Natural Resources throughout the World. between households within a community) can be a valuable IIED and IUCN. Available at www.iapad.org/sharing_power. way to collect data on which species are being harvested from a htm wetland. Market surveys can also provide a key opportunity for Bridson, D. and Forman, L. (eds.). 2004. The Herbarium the fi eld survey teams to integrate their work and obtain cross- Handbook. Third Edition. Kew Publishing, London, UK. cutting data on livelihood and economic values. CBD. 2006. Guidelines for the rapid ecological assessment of biodiversity in inland water, coastal and marine areas. CBD It is important to consider potential biases in market surveys. Not Technical Series No. 22 / Ramsar Technical Report No. 1. all species or products are openly traded, either because they Joint publication of the Secretariat of the Convention on or their trade are illegal, or because their trade takes place in Biological Diversity, Montreal, Canada, and the Secretariat informal markets, perhaps between or within households. Cross- of the Ramsar Convention, Gland, Switzerland. Available at: checking between results from market and household surveys, www.cbd.int/doc/publications/cbd-ts-22.pdf though time-intensive, may show discrepancies in volumes that Cook, C.D.K. 1996. Aquatic Plant Book. SPB Academic reveal undisclosed trade in illegal or threatened species.

CHAPTER 3 47 SECTION II THE TOOLS Edward Allison/Darwin Integrated Wetland Assessment Project Edward Allison/Darwin Integrated Wetland Kong Kim Sreng/Darwin Integrated Wetland Assessment Project Kong Kim Sreng/Darwin Integrated Wetland

Freshwater snails for sale in Stung Treng Ramsar Site

Morning fi sh market in Stung Treng B9.2 Approaches

It is suggested that a fully integrated team of researchers A market survey can provide data on species economic participate in a market survey to ensure that species being traded value, the quantities being traded, the economic status of the can be rapidly identifi ed and that relevant information (purchase households which are harvesting and selling the resource, as and sale prices, volumes, livelihood data on harvesters and well as provide information on institutional factors that impact traders, for example) is collected effi ciently. The data collected upon livelihoods and biodiversity, such as legislation and the should be jointly analysed to provide areas for potential follow- effectiveness of regulation. Surveys can also highlight key on work by the separate researchers, for example: locating and areas of conservation concern, such as the trade in threatened georeferencing harvesting areas, such as favoured fi shing areas; species or their parts, as well as drivers of trade (food, fuel or identifying species and assessing their conservation status; construction, medicinal use, the national or international pet or identifying the wealth class of harvesters and traders, and so on. aquarium trade and so on). Seasonality is likely to be important in planning the timing Market surveys have been used extensively by researchers of market surveys to ensure that seasonal fl uctuations in investigating the extent and impact of trade in wild species, and availability, harvesting levels, and prices due to climate and their reports provide good case studies for undertaking market species migrations are taken into account. Similarly, some surveys (for example, see Singh et al. 2006). products are traditionally sold at particular times of day (e.g.

Table 7: Suggested monitoring schedule for selected sites in Stung Treng town, Stung Treng Province, Cambodia (from Bezuijen et al. 2005)

DAY MORNING MORNING LATE AFTERNOON OTHER (0600-0800) (1100-1400) (1400-1800) (IF TIME) 1 1. Survey food section – Visit Sokh Pheaph 1. Survey jewellery vendors at town markets. town market. restaurant for lunch 2. Monitor river road (town-6 km east to 2. Enter data. airport) for wildlife transport. 3. Enter data 2 1. Survey food section – Visit Sunntha restaurant 1. Monitor river road (town-3.5 km west to 0500-brief town market. (near market) for lunch boat landing) for wildlife transport. survey of main 2. Enter data. 2. Enter data. boat landing 3 1. Survey food section – Visit Prachum Tonle 1. Monitor river road (town-6 km east to Visit Sunntha town market. restaurant for lunch airport) for wildlife transport. Restaurant (2nd 2. Enter data. 2. Enter data. outlet) for dinner 4 1. Survey food section – Visit any new food outlet 1. Monitor river road (town-3.5 km west to 0500-brief town market. (random survey) boat landing) for wildlife transport. survey of main 2. Enter data. 2. Enter data. boat landing

48 CHAPTER 3 THE TOOLS SECTION II

Table 8: Degradation and deterioration of habitats and ecosystems (qualitative/quantitative)

CONTRIBUTORY FACTORS METHODS OF VERIFICATION (INDICATORS) LINKS TO DRIVING INDICATORS

Loss/degradation of wetlands: Increased demand for land; expansion of Landfi ll (area); drainage activities reclamation, drainage agricultural land

Dead/dying aquatic organisms. Pollution of water from Mis-use/over-use of agrichemicals; agrichemicals (fertilizers, Eutrophic conditions – growth of algal mats pesticides etc.) and other effl uents harmful practices related to handling/ Decline in aquatic species’ abundance or application of agrichemicals (oil etc.) distributions

Agricultural activities (i.e. river bank Clearance of riparian vegetation Area of riparian vegetation cleared cultivation use

Upstream dams, diversions etc. (related reductions in water levels) Demand for irrigation water and energy Regulation of water fl ow Extraction of surface or groundwater for (hydropower) agriculture, industry, or domestic use

Ribbon development (settlements etc.) Waste disposal Area of waste dumps bordering wetlands

Table 9: Spread of invasive alien species

CONTRIBUTORY FACTORS METHODS OF VERIFICATION (INDICATORS) LINKS TO DRIVING INDICATORS Deliberate and/or accidental Presence and distribution/spread of invasive alien Expansion of agriculture, aquaculture, introduction of invasive alien plants plant and animal species ornamental fi sheries etc. and animals Decline in native species’ abundance or distribution Loss or degradation of habitats or ecosystem function

fi sh) and such variations also need to be considered. See Table Key threats can be mapped and presented in the project GIS 7 for an example of a sampling programme undertaken as part (see M9). of a survey of markets and outlets in Stung Treng, Cambodia (taken from Bezuijen et al. 2005). B11 Assessing the conservation status of species B10 Assessing threats to freshwater species and ecosystems Conservation actions are often based on the location of threatened species and determining the conservation status Information on threats and changes to the wetland environment of species within the integrated assessment study area will as well as evidence of change in the health, abundance, potentially have a signifi cant impact on the fi nal analysis and and distribution of wetland species is required to inform the recommendations arising from the integrated assessment assessment of the conservation status of wetland species (see process. B11), and to gain an understanding of processes and drivers of change within the survey area. A vital part of the integrated assessment process will be to identify species present within the study area through the Information on degradation and deterioration of habitats and biodiversity, livelihoods, and economic valuation fi eldwork and ecosystems (Table 8), spread of invasive alien species (Table to ascertain their conservation status if possible. The IUCN 9), and on over-exploitation and destruction of species (Table Red List is widely recognised as an independent measure 10) can be collected by direct observation during the course of of a species’ conservation status and this is preferred where biodiversity survey, as well as by focus group and key informant the species has already been assessed against the Red List interviews, and through market surveys (see B9). Criteria. Undertaking a comprehensive conservation status

CHAPTER 3 49 SECTION II THE TOOLS

Table 10: Over-exploitation and destruction of species

CONTRIBUTORY FACTORS METHODS OF VERIFICATION (INDICATORS) LINKS TO DRIVING INDICATORS Illegal poaching of animals (birds, mammals, reptiles etc.) Animals displayed for sale in local markets; traps Unsustainable harvesting observed; presence of hunters; information from Demand for bush meat Harvesting and trade of locals; charcoal transport and sale etc. endangered species

Harmful fi shing and harvesting Blast fi shing, poisoning, electro-fi shing etc. Demand for fi sh practices

Logged areas; log sawing pits; transit timber Logging (riparian trees) Demand for timber depots; timber products etc

Collection of plants and animals Demand for animals and plants in the for ornamental purposes Collections observed; specimens in local markets ornamental trade (commercial trade)

Wanton/deliberate killing e.g. Information from local communities Fear/mythical beliefs reptiles

assessment of all species within a wetland using the IUCN Red There are nine Categories in the IUCN Red List system: List Criteria is likely to be beyond the scope of an integrated Extinct, Extinct in the Wild, Critically Endangered, Endangered, wetland assessment project, and where this is the case, the Vulnerable, Near Threatened, Least Concern, Data Defi cient, following alternatives can be considered. and Not Evaluated, and a further two Categories that are used at the regional scale: Regionally Extinct and Not Applicable. Three approaches can be taken, depending on the resources Classifi cation into the Categories for species threatened with available: (i) search the existing global and sub-global (national extinction (Vulnerable, Endangered, and Critically Endangered) and regional) Red Lists to see if the species present in the is through a set of fi ve quantitative Criteria that form the heart wetland have already been assessed (see B11.6); (ii) select of the system (Figure 19). These Criteria are based on biological a small number of key species that are known to be vital to factors related to extinction risk and include: rate of decline, local livelihoods, such as some fi sh species, and use the population size, area of geographic distribution, and degree of available data to assess their conservation status against population and distribution fragmentation. the IUCN Criteria. Once data have been collated and entered into the SIS database, the threatened status of each species B11.2 The Red List categories and their application can be assessed according to the IUCN Red List Categories and Criteria (see www.iucnredlist.org); and (iii) use anecdotal EXTINCT (EX): A taxon is Extinct when there is no reasonable information, for example, provided by key informants or doubt that the last individual has died (that is, when exhaustive through focus group discussions to indicate historical declines surveys in known and/or expected habitat, at appropriate times in a species’ abundance or distribution within the assessment (diurnal, seasonal, annual), throughout its historic range have area, as well as the causes of the change. In each case, the failed to record an individual). Surveys should be over a time SIS (see Chapter A9.1) database can be used to collate data frame appropriate to the taxon’s life cycle and life form. from the biodiversity, livelihoods, and economic elements of the assessment. EXTINCT IN THE WILD (EW): A taxon is Extinct in the Wild when it is known only to survive in cultivation, in captivity or as B11.1 The IUCN Red List of Threatened Species a naturalized population (or populations) well outside the past range. The Red List, in conjunction with the comprehensive data compiled to support it, has become an increasingly powerful REGIONALLY EXTINCT (RE): Taxa that are considered extinct tool for conservation planning, management, monitoring, and within the region but populations still exist elsewhere in the decision-making (e.g. Rodrigues et al. 2006). world.

50 CHAPTER 3 THE TOOLS SECTION II

(a) Extinct (EX) (b) CRITERIA Extinct in the Wild (EW) Population Regionally Extinct (RE) reduction Threatened categories Critically Endangered (CR) Extinction Restricted Endangered (EN) Risk geographic range Vulnerable (VU)

Small population QUANTITIVE RED LIST Near Threatened (NT) size & decline THRESHOLDS CATEGORIES Least Concern (LC) Very small or restricted All species Data Defi cient (DD) population Not Applicable (NA) Quantitive Not Evaluated (NE) analysis

Figure 19: IUCN Red List (a) Regional Categories and (b) Criteria

CRITICALLY ENDANGERED (CR): A taxon is Critically status. A taxon in this Category may be well studied and its Endangered when the best available evidence indicates that biology well known, but appropriate data on abundance and/ it meets any of the Criteria A to E for Critically Endangered or distribution are lacking. Data Defi cient is therefore not a (see Red List Categories and Criteria booklet (IUCN 2001) for category of threat. Listing of taxa in this Category indicates that details) and it is therefore considered to be facing an extremely more information is required and acknowledges the possibility high risk of extinction in the wild. that future research will show that a threatened classifi cation is appropriate. It is important to make positive use of whatever ENDANGERED (EN): A taxon is Endangered when the best data are available. In many cases great care should be exercised available evidence indicates that it meets any of the Criteria A in choosing between DD and a threatened status. If the range to E for Endangered and it is therefore considered to be facing of a taxon is suspected to be relatively circumscribed, and a a very high risk of extinction in the wild. considerable period of time has elapsed since the last record of the taxon, threatened status may well be justifi ed. VULNERABLE (VU): A taxon is Vulnerable when the best available evidence indicates that it meets any of the Criteria A NOT APPLICABLE (NA): Taxa that have not been assessed to E for Vulnerable and it is therefore considered to be facing a because they are unsuitable for inclusion in a regional Red high risk of extinction in the wild.

NEAR THREATENED (NT): A taxon is Near Threatened when it has been evaluated against the Criteria but does not qualify Jens Kipping for Critically Endangered, Endangered or Vulnerable now, but is close to qualifying for, or is likely to qualify for, a threatened Category in the near future.

LEAST CONCERN (LC): A taxon is Least Concern when it has been evaluated against the Criteria and does not qualify for Critically Endangered, Endangered, Vulnerable or Near Threatened. Widespread and abundant taxa are included in this Category.

DATA DEFICIENT (DD): A taxon is Data Defi cient when there is inadequate information to make a direct, or indirect, assessment Notiothemis robertsi, a widespread species of dragonfl y from central of its risk of extinction based on its distribution and/or population Africa

CHAPTER 3 51 SECTION II THE TOOLS

Table 11: Summary of the fi ve Red List Criteria (A–E) used to evaluate if a taxon belongs in a threatened Category (Critically Endangered, Endangered or Vulnerable). This Summary needs to be used in connection with a thorough understanding of the full Red List Guidelines (IUCN 2001)

Use any of the Criteria A–E Critically Endangered Endangered Vulnerable

A. Population reduction Declines measured over the longer of 10 years or 3 generations A1 > 90% > 70% > 50% A2, A3 & A4 > 80% > 50% > 30% A1. Population reduction observed, estimated, inferred, or suspected in the past where the causes of the reduction are clearly reversible AND understood AND have ceased, based on and specifying any of the following: (a) direct observation (b) an index of abundance appropriate to the taxon (c) a decline in area of occupancy (AOO), extent of occurrence (EOO) and/or habitat quality (d) actual or potential levels of exploitation (e) effects of introduced taxa, hybridization, pathogens, pollutants, competitors or parasites. A2. Population reduction observed, estimated, inferred, or suspected in the past where the causes of reduction may not have ceased OR may not be understood OR may not be reversible, based on (a) to (e) under Al. A3. Population reduction projected or suspected to be met in the future (up to a maximum of 100 years) based on (b) to (e) under Al. A4. An observed, estimated, inferred, projected or suspected population reduction (up to a maximum of 100 years) where the time period must include both the past and the future, and where the causes of reduction may not have ceased OR may not be understood OR may not be reversible, based on (a) to (e) under Al. B. Geographic range in the form of either B1 (extent of occurrence) AND/OR B2 (area of occupancy) B1. Extent of occurrence (EOO) < 100 km² < 5,000 km² < 20,000 km² B2. Area of occupancy (AOO) < 10 km² < 500 km² < 2,000 km² AND at least 2 of the following: (a) Severely fragmented, OR Number of locations = 1 < 5 < 10 (b) Continuing decline in any of: (i) extent of occurrence; (ii) area of occupancy; (iii) area, extent and/or quality of habitat; (iv) number of locations or subpopulations; (v) number of mature individuals. (c) Extreme fl uctuations in any of: (i) extent of occurrence; (ii) area of occupancy; (iii) number of locations or subpopulations; (iv) number of mature individuals. C. Small population size and decline Number of mature individuals < 250 < 2,500 < 10,000 AND either C1 or C2: C1. An estimated continuing 20% in 5 years or 2 10% in 10 years or 3 25% in 3 years or 1 generation decline of at least: generations generations (up to a max. of 100 years in future) C2. A continuing decline AND (a) and/or (b): (ai) Number of mature individuals in each < 50 < 250 < 1,000 subpopulation: OR OR (aii) % individuals in one 90–100% 95–100% 100% subpopulation = (b) Extreme fl uctuations in the number of mature individuals. D. Very small or restricted population Either: Number of mature individuals < 50 < 250 D1. < 1,000 AND / OR D2. Typically: AOO < 20 km² or Restricted area of occupancy number of locations <5 E. Quantitative Analysis Indicating the probability of > 50% in 10 years or 3 > 20% in 20 years or 5 > 10% in 100 years extinction in the wild to be: generations (100 years max.) generations (100 years max.)

52 CHAPTER 3 THE TOOLS SECTION II

NO 1. Assess regional 2c. Is the immigration Downgrade population according to expected to decrease? category from the Red List Criteria YES/ step 1 DO NOT KNOW YES YES 2d. Is the regional Upgrade category population a sink? from step 1

2a. Is the taxon a NO 2b. Does the regional NO/DO non-breeding population experience any NOT KNOW No change from visitor? DO NOT signifi cant immigration of KNOW propagules capable of step 1 reproducing in the region? NO/DO NOT KNOW YES

2g. Can the YES Downgrade 2e. Are the conditions NO 2f. Are the conditions NO breeding category from outside the region within the region population rescue step 1 deteriorating? deteriorating? the regional population should YES/ YES/ it decline? DO NOT DO NOT KNOW KNOW NO/ DO NOT KNOW No change from step 1

Figure 20: Conceptual scheme of procedure for assigning IUCN Red List category at the regional level

List (e.g. a taxon that occasionally breeds in the region under level. Certain adjustments are made to the methods used for favourable circumstances but regularly becomes regionally global assessments, but the process is otherwise the same. Two extinct; see the Guidelines for Application of IUCN Red List additional categories are included for regional assessments: RE Criteria at Regional Levels (IUCN 2001) for other examples of and NA (see Chapter B11.3 above). when this category might be used). Figure 20 shows a conceptual scheme of the procedure for NOT EVALUATED (NE): A taxon is Not Evaluated when it is has assigning an IUCN Red List Category at the regional level. In not yet been evaluated against the Criteria. Step 1 all data used should be from the regional population, not the global population. The exception is when evaluating B11.3 The Red List process a projected reduction or continued decline of a non-breeding population; in such cases conditions outside the region must The process of Red Listing involves compiling data on a species be taken into account in Step 1. Likewise, breeding populations (either globally or within a defi ned region) and then assessing may be affected by events in, for example, wintering areas, that species against a set of criteria to predict the risk of that which must be considered in Step 1. species going extinct. This process is described in detail in the Red List Categories and Criteria booklet (IUCN 2001) and a one- In Step 2 various conditions relating to external factors affecting page summary of the Criteria used for the threatened Categories the population (e.g. immigration) are evaluated to decide whether is also available (see Table 11). To summarise, each species is to ‘upgrade’ or ‘downgrade’ the assigned Red List Category. If the assessed against each of the criteria A-E. The fi nal category of regional population is a demographic ‘sink’ and the extra-regional threat to the species is then determined as the highest level of source population is expected to decline, the preliminary Category threat assigned under any of the criteria. from Step 1 may be upgraded (i.e. EN upgraded to CR; VU upgraded to EN; NT upgraded to VU). If the regional population experiences B11.4 Regional assessments a ‘rescue effect’ (from an external demographic ‘source’) through immigration from outside the region, the preliminary Category The Red List Criteria were initially developed for application from Step 1 may be downgraded (i.e. CR downgraded to EN; at the global scale (i.e. to assess the global population of a EN downgraded to VU; VU downgraded to NT). Other categories species). Red-listing (the process of assessing the conservation (EX, EW, RE, DD, NA, NE and LC) cannot be downgraded or status of a species using the Red List Criteria) is also possible upgraded. Importantly, if a species being assessed is endemic to at smaller scales; at the regional, national and sub-national the assessment region, no regional adjustments are required.

CHAPTER 3 53 SECTION II THE TOOLS

therefore be useful to know if such species are present. The IUCN Red List website (www.iucnredlist.org) should be consulted to see if the species present in the site have already been assessed as threatened at either regional (see the ‘Initiatives’ tab on the main page of the Red List website) or global scales. In many cases the Red List status of species present will not yet have been assessed and the surveyors may consider conducting their own Red List assessments for those species. The feasibility and benefi t of assessing the conservation status of species present within the assessment site is highly dependent upon: i) the size of the wetland site, and ii) the proportion of the species’ total population that is

Kong Kim Sreng/Darwin Integrated Wetland Assessment Project Kong Kim Sreng/Darwin Integrated Wetland restricted to that site. If the site is very small and contains only a small proportion of the species’ global population then it is not considered practical to conduct a Red List assessment at that scale. If, however, the site is reasonably large (a large river or lake catchment for example) then it might be useful to conduct an assessment to determine the species’ risk of extinction within that catchment. The selection of species to assess might therefore be based on criteria such as: i) those species with a high value to local livelihoods or ii) species Hylarana macrodactyla, a widespread frog from southeast Asia that endemic to the assessment area. Such an exercise will help to has been assessed as Least Concern according the IUCN Red List. highlight the presence of any species at risk of local or global The species was found during the Stung Treng integrated wetland extinction. assessment An increasing number of freshwater species are being assessed at the global scale, and assessments are also being See the Guidelines for Application of IUCN Red List Criteria undertaken at the regional and national scales. All species of at Regional Levels (IUCN 2001) for further details on the birds are assessed on a regular basis by BirdLife International, procedures to follow, especially for Step 2. and global assessments of all freshwater crabs, mammals and amphibians have been completed. Global assessments are B11.5 Applying the Red List categories to wetland planned or under way for reptiles, dragonfl ies and damselfl ies, species freshwater fi shes, and molluscs, as well as some plant groups (cycads and conifers, for example). All global assessments can The Red List methodologies were designed to be applicable be downloaded from the Red List website (www.iucnredlist. to all species but in practice certain adaptations are necessary org), which can be searched using a range of criteria including when assessing riverine species and creating their distribution taxonomy, location, habitat and system (freshwater, marine or maps. For example, the distribution range of a wetland species, terrestrial). In addition, species are increasingly being assessed an important criterion used to assign a Red List Category, is at the sub-global (regional, national and even sub-national) often calculated to include the total surface area of water and scale (see Chapter B11.4) through a range of processes, for land within the associated river or lake catchments where the instance: species has been recorded, even though it is clear that the ● regional assessment projects undertaken by IUCN (e.g. species is not found on land. In order to take account of such the IUCN Pan-Africa Freshwater Biodiversity Assessment issues a document has been prepared to guide assessors of project and the IUCN European Mammal Assessment) freshwater species (contact the IUCN Freshwater Biodiversity ● regional and national assessments undertaken by national Unit, Cambridge, UK for more information). governments, academic institutions or NGOs (e.g. ZSL National Redlist project http://regionalredlist.com/) B11.6 Assessing the conservation status of species during integrated wetland assessments Some of the regional assessment data, for example from the Pan-Africa freshwater assessment (see www.iucn.org/species/ The presence of threatened species within a wetland site freshwater) and similar forthcoming initiatives in Asia will be may have signifi cant implications for decisions on the future available through the Red List website (www.iucnredlist.org; conservation or development of a wetland site. It would see the ‘Initiatives’ tab).

54 CHAPTER 3 THE TOOLS SECTION II

BOX 6: PARTICIPATORY RESEARCH ON FISH SPECIES conservation approaches to approaches with a broader focus AND FISH-BASED LIVELIHOODS on environmental conservation (Pickett et al. 1997).

The Mekong Wetlands Biodiversity Programme worked with The choice of conventional species-based measures of local villagers to document fi sh species and fi shery-related diversity as advocated in this toolkit has both advantages issues in the Lower Songkhram River Basin in Thailand. and disadvantages. The main advantages include that the People from four villages took part between May 2003 and results will be comparable with past and future surveys of April 2005. Within the fl ooded forest on the river fl oodplain, the same type and that the survey outputs are likely to be Thai Baan researchers identifi ed 208 types of vegetation and broadly acceptable to a wide range of people. Importantly, the fungi that local people consume or use. Twenty-eight types species-based approach makes it possible to link with Red of riverine sub-ecosystems were distinguished according to local terminology, many of which are important fi sh habitats List procedures which currently provide the basis for much particularly for spawning. One hundred and twenty four conservation planning. species of fi sh, six species of turtle, four species of shrimp, 10 species of molluscs and four species of crabs were The disadvantage of using conventional taxonomic-based identifi ed and photographed, and notes were made on their measures of biodiversity include the limited knowledge of ecology, such as whether they migrate, how far they migrate taxonomy of many poorly-studied taxonomic groups, and the and when. The researchers also considered the status of scarcity of taxonomic experts; these knowledge gaps constrain fi sh species, noting that 14 species are now rarely caught the range of taxa that can be chosen for survey. (considered ‘endangered’) and 12 species are no longer seen, and likely to be locally extinct. Local people are uniquely A number of alternative approaches have been developed for placed to collate this information, as they adapted their use by major biodiversity projects. livelihoods over many years to utilize the fi sh resources based on a deep understanding of fi sh migration patterns, feeding B12.1 Parataxonomy and spawning, fl ood patterns and fi sh habitats. The use of non-specialist technicians as parataxonomists to distinguish ‘morphologically recognisable taxonomic units’ (Oliver and Beattie 1993, 1996a, 1996b) for sorting large Advice on submitting assessments, as well as the information samples. Expert time is expensive and there are not enough required and the format of assessments can be obtained from experts available to carry out the large amount of routine the IUCN Red List Unit ([email protected]). sample processing required by major biodiversity surveys. Trials with insect species showed that with a few hours’ training, B11.7 Key references non-specialist technicians and students performed with 87% accuracy compared to formally trained taxon-specialists (Oliver IUCN. 2001. Red List Categories and Criteria (Version 3.1). and Beattie 1993). This level of accuracy is likely to suffi ce for IUCN Red List Unit, Cambridge, UK. purposes of conservation management, where error variances IUCN. 2008. Guidelines for Using the IUCN Red List and bias associated with sampling techniques are likely to Categories and Criteria. Version 7.0. Prepared by the over- or under-estimate species’ richness by greater margins. Standards and Petitions Working Group of the IUCN SSC Most major biodiversity projects in species-rich rainforests Biodiversity Assessments Sub-Committee in August 2008. make extensive use of large numbers of parataxonomists (for Downloadable from http://intranet.iucn.org/webfi les/doc/ example Tangley 1990, Cranston and Hillman 1992, Kaiser SSC/RedList/RedListGuidelines.pdf 1997).

B12 Alternative methods for biodiversity B12.2 Participatory biodiversity assessment and assessment monitoring

While species-based methods of assessment are widely Wetland resource users generally have a great deal of non- used and accepted, they also encounter diffi culties – such scientifi c indigenous knowledge about their environment and the as the lack of available taxonomists, problematic defi nitions species in it. Involving people living in wetlands in biodiversity of species, and even the species concept itself (e.g. Mishler assessment and monitoring has advantages besides being a and Donoghue 1982, Turner 1999, Wheeler and Meier 2000). cost-effective use of existing information: Species diversity may not be the most important diversity- ● it minimises the requirements for expensive and sometimes related attribute of an ecosystem (Bengtsson 1998, Schwartz distant expert input et al. 2000), leading some to move away from species-based ● it involves local resource-users, who have a larger stake in

CHAPTER 3 55 SECTION II THE TOOLS

the future of the resources than any government offi cial or techniques, which employ some of the approaches outlined visiting scientist above, vary in their data requirements, cost, and suitability ● it serves to maintain dialogue and build co-operative for application for different purposes and at different spatial understanding between local stakeholders, resource users, scales. This Integrated Wetland Assessment methodology is researchers, and resource managers similar to the Rapid Assessment Programme (RAP) developed by Conservation International for surveys of poorly-known The importance of using indigenous understanding of natural areas, using ‘surrogate’ or ‘indicator’ groups identifi ed to resource systems to assess, manage and monitor natural species level by small teams of national and international resources, including biodiversity, is now widely recognised experts (see Groombridge and Jenkins 1996). These surveys (see for example Hellier et al. 1999 and a review by Sillitoe are then used to assess conservation value by assuming a 1998) beyond the boundaries of ethnobotany where it has relationship between these ‘indicator’ groups and total diversity long been a legitimate research method (Martin 1995). This and habitat quality. The main drawbacks of the methodology approach has been used in the Lower Songkhram River are the reliance on specialist taxonomic expertise (beyond Basin in Thailand (Thai Baan; see Box 6), as well as during standard fi eld identifi cation skills) and the assumptions made the integrated wetland assessment undertaken in Stung Treng about relationships between indicator diversity and total Ramsar Site (see Chapter 7) where the methodology has been diversity. called Sala Phoum, or village research. Other rapid assessment methods include Conservation B12.3 The use of higher-taxon approaches Biodiversity Workshops, Conservation Needs Assessments, Gap Analysis and Biodiversity Information Systems If the hierarchical taxonomic classifi cation system has any (Groombridge and Jenkins 1996). Some of these methods do objective validity, then it is obvious that higher levels of taxa not require additional survey work and aim to make best use of provide integrative summaries of diversity within each level of existing information, including socio-economic data that can classifi cation. In principle, any level of taxonomic classifi cation be overlooked by biodiversity specialists. can be chosen for comparative analysis. By convention the species level is chosen, but where identifi cation to species is B12.5 Key resources not possible it is common to use higher-taxon approaches. There is some experience indicating that correlation between Conservation International’s Center for Applied Biodiversity diversity at different taxonomic levels can be established Science – Rapid Assessment Programme http://science. (Balmford et al. 1996), although this is likely to be highly conservation.org/ variable (Gaston and Williams 1993, Prance 1994, Williams Groombridge, B. and Jenkins, M.D. 1996. Assessing and Gaston 1994, Anderson 1995). Balmford et al. (1996) biodiversity status and sustainability. WCMC Biodiversity found that using woody plant genera and families, rather than Series No 5. World Conservation Press, Cambridge, UK. species, yielded comparable estimates of relative conservation value of tropical forest for 60–85% less cost than a species- Further reading based survey. It may be possible to use a much wider range of taxa, for lower sample processing effort, if the principle of CBD. 2006. Guidelines for the rapid ecological assessment of higher-taxon comparisons proves acceptable. Biotic indicators biodiversity in inland water, coastal and marine areas. CBD of ecosystem health (which should be related to diversity) in Technical Series No. 22 / Ramsar Technical Report No. 1. aquatic systems are usually based on identifi cation of macro- Joint publication of the Secretariat of the Convention on invertebrates to higher taxonomic levels, such as genus or Biological Diversity, Montreal, Canada, and the Secretariat family (Chessman 1995, Hilsenhoff 1988). of the Ramsar Convention, Gland, Switzerland. Available at: www.cbd.int/doc/publications/cbd-ts-22.pdf B12.4 Rapid assessment techniques MEA. 2005. Ecosystems and human wellbeing: Wetlands and water synthesis. World Resources Institute, Washington, In recognition that the task of determining a conservation DC. Available at: http://www.millenniumassessment.org/ strategy is urgent in areas where biodiversity is threatened, documents/document.358.aspx.pdf highly utilized and poorly known, a number of techniques for Ramsar Handbooks for Wise Use. Available from: the Ramsar rapid assessment of conservation value have been developed Wise Use Resource Centre www.ramsar.org/wurc/wurc_ (reviewed in Groombridge and Jenkins 1996). These index.htm

1 Fish Species in the Wetlands of the Lower Songkhram River Basin – Local Knowledge of the Fishers in the Lower Songkhram River Basin. IUCN and WANI. Available in Thai with an English introduction from: www.mekongwetlands.org/Common/download/Thai_Fish_Book_2.pdf

56 CHAPTER 3